Applications · Photovoltaics
Solar cells
Use absolute luminescence to locate voltage, recombination, transport and uniformity losses from absorber films to complete solar cells and modules.
Layer-by-layer loss analysis
How much of the difference between sample voltage and its radiative limit is caused by the bulk absorber, by each interface and by transport to the contacts?
PLQY and QFLS are measured at the 1-sun-equivalent generation rate on a sequence of samples: the neat absorber, the absorber with each transport layer (half-stacks), the full stack and the finished cell. Excitation-intensity sweeps on each sample give pseudo-JV curves. The SMU records the electrical J–V curve of the finished cell. ELQY measured with 1-sun-equivalent injected current on the full stack independently cross-checks losses.
Instrument: LuQY Pro for single sample, LuQY Scanner for all samples in the sequence (+mapping, see here)
What you learn · each stack step isolates one loss
- Neat absorber: non-radiative loss in the bulk and at free surfaces, ΔVnr = −(kBT/q) ln(PLQY), relative to the radiative limit.
- Half-stacks: additional recombination introduced by each interface, such as SAM layers, transport layers or buffer layers/interlayers for tandem cells. The interface that lowers the QFLS most is the first target for passivation [3].
- Finished cell: the difference between QFLS and qVOC, and the difference between the pseudo fill factor and measured fill factor. The latter separates recombination-limited fill factor from transport and series-resistance losses.
- Implied efficiency of the end device: Intensity sweeps on the neat absorber and subsequent stacks also give the implied efficiency: the efficiency a stack or device could reach if carrier collection introduced no further losses [4]. It is defined using iVOC (or pseudo-VOC), the pseudo-FF and the short-circuit photocurrent. The maximum power output is marked by the pseudo maximum-power point. Plotting the pseudo-JV parameters alongside QFLS for each stack gives predictive insight into sequential losses before the device is fully built.
- PLQY of 10-100% → Interpreted as nearly all radiative recombination. Indicates excellent material quality (like GaAs).
- PLQY of 0.1-10% → Increased fraction of non-radiative losses; still allowing for highly efficient solar cells (e.g. world record cells in perovskites).
- PLQY of 0.01 – 0.1% → Significant non-radiative losses; material still usable but improvements recommended.
- PLQY< 0.01% → Dominant non-radiative recombination; poor optoelectronic quality.
The QFLS losses of the individual layers do not simply add up. The QFLS is the logarithm of the recombination current, and is therefore dominated by the strongest recombination channel across the stack. [3] Samples should be compared on the same substrate type and at the same absorbed photon flux.
QFLS is generally an accurate predictor of the external device Voc, but can deviate from this value under strong mismatch of transport layer work functions, low mobility of absorber layers or interlayers, mobile ion & field-screening effects, etc. [2]
For finished devices, electroluminescence provides an independent cross-check. If the cell is driven with an injected current density equal to its JSC, the optoelectronic reciprocity relation [5] gives Eq. (A1) below. Note that this assumes reciprocity which holds when charge collection does not depend strongly on voltage. This should be checked for each device type.
Spatial uniformity and upscaling
Does the whole area of a large film or module reach the quality of the best spot, and where are the losses?
Absolute PL mapping and PL imaging over areas up to 210 × 210 mm, either as one large sample or several smaller samples in one run.
Instruments: LuQY Scanner for large areas and sample throughput; LuQY Pro for single-sample mapping.
What you learn
- Coating gradients, i.e. inconsistencies in film coating
- Local defects and potential shunt channels
- Edge effects, poor passivation and potential source points of sample degradation
- Damage during post-processing, eg. dewetting or damage near module scribe lines
- Intensity sweeps map the pseudo-FF across the film, showing where recombination limits the achievable fill factor before a device is built.
- With upscaling, several samples are measured in one go, enabling statistics and correlation across large data sets.
The intensity gradients from the PL image cannot be used as a heuristic to access spatial variation in QLFS. More accurately, it is extracted from the local PLQY, which can be obtained only with local spectral analysis, as performed by the LuQY systems. This resolves potential spectral shifts across different measurement points. In finished devices with conductive electrodes, lateral currents redistribute voltage, so a local defect such as a shunt can appear as an extended dark region.
Stability: which loss appears first?
When a device or film degrades, does it lose material quality, change composition, or do the interlayers and contacts deteriorate?
Absolute PL spectra are recorded as a time series under continuous illumination at a fixed intensity. For complete devices, the SMU can simultaneously hold open circuit, a fixed voltage near the maximum power point, or a constant injected current.
Instruments: LuQY Pro; LuQY Lifetime for TRPL before and after.
What you learn → signatures differ by mechanism
- PLQY and QFLS fall together: non-radiative recombination is increasing, for example through defect formation or interface degradation.
- Red-shifted emission or an emerging low-energy band: this can indicate light-induced halide segregation in mixed-halide perovskites [6]. Recovery in the dark indicates a reversible fundamental process.
- VOC falls while QFLS stays constant: the losses are located at the contacts rather than in the absorber.
Note light soaking effects on PLQY: Under extended illumination, PL intensity can increase (photobrightening) or decrease (photodarkening), often depending on the atmosphere.
Voltage-resolved absolute PL
Where between open-circuit and short-circuit does a device lose internal voltage, and how efficiently are photogenerated carriers extracted?
The complete cell is illuminated at the 1-sun-equivalent generation rate while the integrated source-measure unit (SMU) sets the terminal voltage, stepping from VOC through the maximum power point to 0 V (short circuit). At each step the absolute PL spectrum is recorded and the QFLS is extracted.
Instrument: LuQY Pro.
What you learn
- At open circuit, comparing QFLS with qVOC shows whether the full internal voltage reaches the terminals. A QFLS clearly above qVOC points to energetic misalignment of transport layers, non-ideal contacts or ionic field screening, among other causes [1, 2].
- At operating voltages, the difference QFLS − qV is the part of the internal voltage that is lost on the way to the contacts. If this difference grows towards the maximum power point, the fill factor is limited by transport and extraction rather than by recombination alone.
- Towards short circuit, efficient extraction lowers the steady-state carrier density in the absorber, so the PL is strongly quenched. A PL signal that remains high at short circuit indicates that a substantial fraction of carriers recombines before it is collected, which is directly relevant to JSC.
When the device delivers current, meaning that the SMU sinks rather than injects current, the extraction of current needs to be considered as a competition to emissive recombination pathways. In this case, the generalized LuQY of Eq. (5) on the absolute PL theory page is dealt with careful consideration.
